compression waves

[John Delacour]

At 5:18 PM +0100 11/23/01, Richard Brekne wrote:

>Outside of the fact that the results of this experiement fit
>point in the exact same direction that Conklins research points to, my initial
>feelings would be to think this does indeed apply. So.... my next step is to
>build a monochord variation of this apparatus to run the experiment in a more
>piano like set of circumstances.
>
>I suspect actually that the same results will show themselves, but 
>we will see.
>I dont really see any reason to think that the double bearing point clamp no
>matter how massive will act as an inhibitant to longitudindal waves continuing
>through the medium (string). After all, these are internal to the 
>string and the
>clamps are on the outside.

I carried out a similar test with a Ø 9.5 brass-sleeved steel stair 
rod about 78 cm long clamped in a vice precisely in the middle for 
about 8 cm.  Provided the rod was gripped centrally, a clear ring of 
ca. 3000 cps was obtained.  Otherwise it was impossible to obtain any 
tone, including the case where the rod was firmly clamped at each end.

The frequency of a piano string removed from one of the bridges and 
pulled taut by hand, is of a quite different order from its frequency 
when held, even at quite low tension between two hard terminating 
bridges.

By the very nature of the oscillating medium (molecules of steel in a 
certain configuration) some disturbance of at least the central part 
of the rod or string is likely to continue past any clamping 
mechanism.  I would visualize a far greater impedance to the wave in 
a 1 mm wire than in a 10 mm rod, though this would need to be tested.

I did the test just out of curiosity but saw no use in it for the 
purposes of our discussion since the conditions in the Kundt 
experiment are so far removed from piano technology.

We are concerned with compression waves in taut steel wire, their 
undoubted existence as audible tones, their undoubted exchange of 
energy with the transverse waves, and the means and quantity of their 
continuance past simple and compound bearings such as are either 
commonly used or such as might be conceived of.  Beyond these 
considerations, except as an aid to understanding basic principles, 
more generalized experimentation is pointless.

The compression wave is generally represented as a series of vertical 
lines oscillating like the coils of a 'Slinky'.  That's fine for 
elementary school but a picture of the relative position of millions 
of molecules in a steel wire, in close contact at numerous points 
with other metallic objects would be very different.  It is almost 
certain that the hardened shell of the drawn steel reacts to the 
shock in a different way from the more crystalline interior of the 
wire and that Young's modulus for the two is different.  I for one 
need to have a far clearer picture in my mind of the internal forces 
at work, and I think the answer to questions such as this might go 
some way to explaining the speaking length problem.

All that is very interesting, and personally I believe in 
understanding my materials to the nth degree so far as I'm able, but 
a line needs to be drawn between scientific knowledge as an aid to 
design and that knowledge for its own sake, which has no more to do 
with piano technology than the last words from the cross.

I have just build a nice little 60 mm trichord with sufficient front 
and back lengths to allow of a good range of tuning.  It is quite 
possible that Duplex Scaling started life in the same way.  Thoeodore 
S. talks of longitudinal vibrations at a time when he would probably 
have had only Lord Rayleigh's work as a reference.  In view of the 
base length of the waves in the upper part of the scale, it is hard 
to see they could have any effect, and yet the tuning and detuning of 
the partials in the duplex scale, not to speak of the damping or not 
of the free wire between hitchpin and brass bridge, have a 
considerable effect on the ringing quality of the heard tone, and to 
dismiss compression waves as beyond range might be simplistic -- a 
reading of Jim Ellis' patent might give some credence to Steinway.

One last unreported fact:  During the few tests I carried out on the 
stringmaking machine, I wondered what would happen if I loosened the 
copper coils to get a buzzing or rattling string.  With the coils 
loose, it was quite impossible to get any semblance of an audible 
compression wave.  This suggests to me initially  not that the loose 
coils prevent the disturbance of the molecules along the steel wire 
but that this disturbance, in order to be audible MUST be converted 
into transverse waves.  Our hearing of a sound is enabled only by 
compression waves in the air acting on the diaphragm of the ear.  A 
wave moving through a steel rod or wire cannot conceivably be audible 
unless it causes a significant movement of air molecules and this it 
will not do, any more than light waves will escape from an optical 
fibre.  The audibility of the compression wave must therefore be due 
to its conversion to transverse movement of the wire which does set 
up compression waves in the air.

My explanation of the loose coil effect is that the normal movement 
of the wire excited by the compression wave is totally absorbed by 
the loose coils.

JD